You are on page 1of 16

This article was downloaded by: [The Library, University of Witwatersrand]

On: 12 February 2015, At: 03:53


Publisher: Routledge
Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered
office: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK

International Journal of Cultural Policy


Publication details, including instructions for authors and
subscription information:
http://www.tandfonline.com/loi/gcul20

Four models of the creative industries


a

Jason Potts & Stuart Cunningham

CCI (ARC Centre of Excellence for Creative Industries and


Innovation) , Queensland University of Technology , Brisbane,
Australia
Published online: 13 Aug 2008.

To cite this article: Jason Potts & Stuart Cunningham (2008) Four models of the creative industries,
International Journal of Cultural Policy, 14:3, 233-247, DOI: 10.1080/10286630802281780
To link to this article: http://dx.doi.org/10.1080/10286630802281780

PLEASE SCROLL DOWN FOR ARTICLE


Taylor & Francis makes every effort to ensure the accuracy of all the information (the
Content) contained in the publications on our platform. However, Taylor & Francis,
our agents, and our licensors make no representations or warranties whatsoever as to
the accuracy, completeness, or suitability for any purpose of the Content. Any opinions
and views expressed in this publication are the opinions and views of the authors,
and are not the views of or endorsed by Taylor & Francis. The accuracy of the Content
should not be relied upon and should be independently verified with primary sources
of information. Taylor and Francis shall not be liable for any losses, actions, claims,
proceedings, demands, costs, expenses, damages, and other liabilities whatsoever
or howsoever caused arising directly or indirectly in connection with, in relation to or
arising out of the use of the Content.
This article may be used for research, teaching, and private study purposes. Any
substantial or systematic reproduction, redistribution, reselling, loan, sub-licensing,
systematic supply, or distribution in any form to anyone is expressly forbidden. Terms &
Conditions of access and use can be found at http://www.tandfonline.com/page/termsand-conditions

International Journal of Cultural Policy


Vol. 14, No. 3, August 2008, 233247

Four models of the creative industries

Downloaded by [The Library, University of Witwatersrand] at 03:53 12 February 2015

Jason Potts* and Stuart Cunningham


CCI (ARC Centre of Excellence for Creative Industries and Innovation), Queensland University of
Technology, Brisbane, Australia
(Received
10.1080/10286630802281780
GCUL_A_328345.sgm
International
1028-6632
Original
Taylor
302008
14
j.potts@qut.edu.au
JasonPotts
00000August
and
&
Article
Francis
00
(print)/1477-2833
Francis
Journal
Month
2008 0000;
of Cultural
in (online)
final
Policy
form 00 Month 0000)

What is the dynamic value of the creative industries from the economic perspective?
This paper seeks to answer this question by proposing four models of the relationship
between the creative industries and the whole economy, then examining the evidence for
each. We find that growth models fit the data well, but not everywhere. We discuss the
methodological and empirical basis for this finding and its implications for economic
and cultural policy.
Keywords: creative industries; economic evolution; growth models; cultural policy

Introduction
Creative industries is a new analytic definition of the industrial components of the economy in which creativity is an input and content or intellectual property is the output.1 This
definition was introduced in the DCMS 1998 template and has been adopted in raft of
mapping documents by other countries also seeking to estimate the size and growth of this
sector and to formulate new policy.2 The creative industries have thus come to be newly
represented as a significant and rapidly growing set of industries;3 an important sector, in
other words, for policy attention.
The ostensible purpose of these mapping documents has been to estimate the significance of the creative industries to the modern economy in order to reorient economic policy
support in accordance with that significance.4 In doing so, however, these studies highlight
an important point: namely that the economic value of the creative industries may extend
beyond just the manifest production of cultural goods or the employment of creative people,
but may have a more general role in driving and facilitating the process of change across
the entire economy, as evidenced by its dynamic parameters and degree of embedding in the
broader economy. Indeed, it may even be the case that the dynamic significance of the
creative industries is greater than their static significance.
This paper seeks to address this question by posing it directly: what is the dynamic relation between the creative industries and the rest of the economy?5 The four models of this
paper are the four possible answers to this question: namely (1) welfare, (2) competition, (3)
growth and (4) innovation. Each of these possibilities parlays into a very different policy
model: in (1) a welfare subsidy is required; in (2), standard industry policy; in (3), investment and growth policy; and in (4), innovation policy is best. Very different policy frameworks thus follow from each of the four basic dynamic models relating the creative
industries to the rest of the economy. This paper will outline these four models and marshal
a sample of existing evidence to begin the process of sorting among them.
*Corresponding author. Email: j.potts@qut.edu.au
ISSN 1028-6632 print/ISSN 1477-2833 online
2008 Taylor & Francis
DOI: 10.1080/10286630802281780
http://www.informaworld.com

Downloaded by [The Library, University of Witwatersrand] at 03:53 12 February 2015

234

J. Potts and S. Cunningham

We begin by outlining the four possible primary relations between the creative industries and the rest of the economy. We explain the relation of each model to different
theoretical foundations, what we should expect to observe if each model were true, and the
appropriate policy framework in each case. Using various mapping documents, we then
undertake an initial consideration of a set of data samples connected to the four models.
However, this paper does not attempt a comprehensive analysis. For that, a much more
rigorous approach to modeling, data and statistical analysis would be required. What we
aim to provide here is only a theory of the classes of models involved and an illustration of
how different sorts of data might be applied to them. This seems to us a necessary first
step (prior to a more rigorous approach) in developing the economics of the creative
industries.
Yet, in doing so, we immediately find at least superficial evidence supporting models
(3) and (4). And while clearly signaling the need for further theoretical and empirical
work, this also points up the potential value of an innovation-based approach to creative
industries and cultural policy. As such, we propose these four models as a starting point
for further discussion of the intersection between economic analysis of the arts and culture
on the one hand and modern analysis of economic growth (and growth policy) on the
other.
Four models of the creative industries
Let the economic value of the whole economy be defined as Y, and the economic value of
the creative industries as CI, affording us the master equation:6
CI = A.Y

In English, this just says that the creative industries comprise some given fraction (A) of all
economic activity. In a static model, this estimate is treated as the significance of the
sector. In Australia, A is estimated at 0.045.7 The estimate of A has been a central output of
the creative industries mapping documents, beginning with DCMS (1998) and since replicated by Australia, NZ and the EU, among others.8 These estimates all find that the creative
industries are indeed economically significant (in the static sense). Furthermore, they are
deemed thus comparable to other high profile sectors in their contribution to income,
employment and trade agriculture, for example, typically has an A value of 0.03. By implication, the creative industries are argued to deserve policy attention (and support) in proportion to that significance.
The problem with this line of reasoning, however, is it has no basis in economic theory.
It is a matter of political expediency to afford an industrial sector policy attention in proportion to the share of income (or jobs, or foreign exchange) it generates, not a matter of
economic logic.9 This is always true in any equilibrium-based (static) argument. Indeed, it
is only when considering the failure of an industry that political and economic significance
align in this static manner (due to the proportionate and ramifying distortions on other
industries). But the creative industries interaction with the aggregate economy is presumed
to be positive, not negative. Yet, if so, then the basic economic-political significance may
no longer hold.
Instead, economic significance needs to be reconstructed. This is what the four models
approach seeks to do by shifting to a dynamic approach to significance. In this approach,
the economic analysis of the relation between an industry sector and the rest of the economy
is instead constructed in terms of the dynamic interrelationship, which we may specify by

Downloaded by [The Library, University of Witwatersrand] at 03:53 12 February 2015

International Journal of Cultural Policy

235

examining the higher order moments of our master equation: specifically, how a change in
CI activity (CI) affects aggregate economic activity (Y).
Our starting axiom is that change in CI affects Y in some way (CI Y). The
four models proposed are the set of possible dynamic interrelations in which a change
in CI activity has either a negative (model 1), neutral (model 2) or positive (models 3
and 4) effect on the economy. This is plainly simplistic, yet we suggest it offers a
useful starting point to orient both empirical analysis and policy discussion in order to
be clear and explicit about this hypothesized relation and its evidential support.
For analytic convenience, we also assume that dCI/dY = 0, meaning that economic
growth affects the creative industries no differently from other industries, or technically that
income elasticity is unitary.10 We strongly suspect this not to be the case, but rather that
growth in income disproportionately effects demand for the output of the creative industries,
but we shall set that aside here.11 Policy is analyzed in terms of whether change in the
creative industries changes aggregate utility welfare (or utility, U). Again, we presume dU/
dCI can increase, decrease or leave utility unaffected. This is also a highly abstract modeling
formulation, yet it enables us to abstractly discriminate between basic differences in theoretical assumptions and policy responses.

Model 1: The welfare model


In this model, the creative industries are hypothesized to have a net negative impact on the
economy, such that they consume more resources than they produce. A dynamically equivalent statement is that the rate of total factor productivity (TFPCI) growth is less in the
creative industries than in other sectors (TFPY), as assumed in Baumol and Bowen (1966).
In this model, the creative industries are essentially a merit good sector that produces
cultural commodities that are welfare enhancing (dU/dCI > 0), but that are only economically viable with a transfer of resources from the rest of the economy (dY/dCI < 0). Furthermore, positive knowledge spillovers associated with production that would augment TFPY
are excluded.
Hypothesis 1:

dY
dU
< 0,
>0
dCI
dCI

In model 1, the creative industries are a net drain on the economy, although a net drain
worth having, as the overall effect is welfare positive. This is due to the production of
commodities of high cultural value (dU/dCI > 0) but low market value (dY/dCI < 0), as
production is inherently unprofitable because demand curves lies everywhere below cost
curves. The economic justification for such restitution must ultimately then rest on a market
failure argument, with policy appropriately calibrated to estimates of this non-market value.
Yet the question of whether market failure is an appropriate justification for intervention
need not concern us here, for it is sufficient to recognize that if dY/dCI < 0 then policy intervention can be justified only if it is also true that dU/dCI > 0. If model 1 is true, then policy
prescriptions should centre about income and resource reallocation or price maintenance in
order to protect an inherently valuable asset (i.e. cultural production) that is naturally and
continually under threat in a market economy.12
It is broadly accepted by scholars of cultural economics (e.g. Throsby and Withers 1979,
Throsby 1994, 2001) and supported by numerous non-market valuation studies (see, e.g.,
Towse 1997, 2003) that dU/dCI is, on the whole, mostly positive. This is an unsurprising
and indeed edifying result that accords with intuition. It is, furthermore, not inconsistent

Downloaded by [The Library, University of Witwatersrand] at 03:53 12 February 2015

236

J. Potts and S. Cunningham

with economic conceptions of rational economic man (Frey 2003, Dopfer 2004). So, let us
take that positive sign as evidential and focus instead on the sign of dY/dCI and what it
means to say that dY/dCI is negative.
Specifically, it means that growth in the creative industries comes at the cost of aggregate economic growth, as their growth is not what the market wants, but must be
compelled to support it through transfers. Evidence for model 1 may therefore accrue in
several ways. These include: high levels and rates of negative profit among creative industries firms; low total factor productivity (TFPCI < TFPY); persistently lower income to
factors of production in creative industries compared with other industries; and other indications that the economic viability of activities and organizations within the creative
industries is critically dependent upon resource transfers from the rest of the economy to
maintain prices, demand or supply.
If model 1 is true, we would expect to observe not just an economically stagnant or lowgrowth sector, but also one with lower performance levels (e.g. return on investment,
incomes, etc). Such decomposition allows multiple opportunities for empirical assessment.
This is, we think, an interesting empirical question to re-ask, for the implicit truth of model
1 is almost axiomatic in the field of cultural economics where few dispute the implicit assumption of below-average income or productivity growth.13 This differs from the implicit assumption of competition in neoclassical microeconomics, which is instead presumed in model 2.

Model 2: The competition model


Model 2 differs from model 1 in allowing that the creative industries are not economic
laggards, nor providers of special goods of higher moral significance, but effectively just
another industry: in effect, the entertainment or leisure industry. In this model, which is the
default setting in standard microeconomic analysis, a change in the size or value of the
creative industries has a proportionate (but structurally neutral) effect on the whole economy. This model also presumes that the growth impact is also neutral, such that the creative
industries would in aggregate contribute no more or less to technological change, innovation
or productivity growth than the average of other sectors.14
This model does not argue that the creative industries have no effect on income, productivity or welfare, as that is trivially false, but that their effect is on par with all other sectors
such that TFPCI = TFPY. Indeed, this is what standard economic analysis would predict as
based on the competitive substitution of resources in a market-based economy to achieve
equivalent returns at the margin. In other words, standard economic theory predicts model
2, in which the creative industries are normally competitive.
If so, this implies that the marginal welfare benefit of policy-based redirection of
resources into this sector is zero in aggregate. That is, there are no economic welfare gains
to special policy treatment. This implicitly supposes that cultural/creative goods are normal
goods, in the sense that as they vary in relative price, rational consumers would substitute
between them and those from other sectors to equalize their marginal utility. In this case, an
expansion of the creative industries sector would have no aggregate welfare benefit distinct
from expansion of any other sector.
Hypothesis 2:

dY
dU
= 0,
=0
dCI
dCI

Model 2 does not exclude the possibility that the economics of the creative industries
are special in terms of extreme levels of demand uncertainty, power-law revenue models,

Downloaded by [The Library, University of Witwatersrand] at 03:53 12 February 2015

International Journal of Cultural Policy

237

tendencies toward monopoly, complex labour markets and property rights, endemic hold-up
problems, information asymmetries, highly strategic factor markets, and so on (e.g. Caves
2000, De Vany 2004). Rather, it emphasizes that these coordination problems are eventually
solved under competitive conditions, just as the special circumstances of other industries led
them to discover specific institutional arrangements and coordination structures.
Model 2 emphasizes these as problems for management as well as opportunities for
entrepreneurs, but ultimately insists that they are no different to the special problems of
all other industries, such as energy or tourism, which also have interesting features associated with scale, coordination, uncertainty, networks, and so forth. The normal model
thus finds that the creative industries have comparable industry statistics to other sectors.15
It follows, then, that they should properly require the same policy treatment as other industries. The creative industries, in this view, are just another member of the industrial community, and they should rightfully then demand neither more nor less assistance than that due
to others. Recognition of normal existence is sufficient and significance is immaterial.
If so, the creative industries then require no special policy treatment, just the consistent
application of policy mechanisms extended to other industries. This is as if the WTOs
most favoured nation principle were applied to most favoured industry, such that any
privileges extended to one industry must automatically extend to all. In this view, the
creative industries policy focus should not be about resource reallocation, but rather with
the plea for consistent industrial policy treatment (and especially with respect to international movement of labour and intellectual property).
Evidence of the normal model would come from the equivalence of creative industries
economic indicators with those of the whole economy in the form of evidence of normal
competition and enterprise. For the more industrially mature parts of the creative industries,
such as film, TV and publishing, this is generally true, as the dominant firms in these sectors
have experienced relative stability and consolidation over several decades. However, there
is a range of new media which do not fit this pattern and this is the basis of the third model
in which the creative industries facilitate economic growth.
Model 3: The growth model
Model 3 explicitly proposes a positive economic relation between growth in the creative
industries and growth in the aggregate economy, such that dY/dCI > 0. In this model the
creative industries are a growth driver in the same way that agriculture was in the early
twentieth century, elaborately transformed manufacturing was in the 1950s1960s, and ICT
was through the 1980s1990s. There are many possible explanations, but all are some variation upon either the notion that the creative industries introduce novel ideas into the economy
that then percolate to other sectors (e.g. design-led innovation), or that the creative industries
facilitate the adoption and retention of new ideas or technologies in other sectors (e.g. ICT).
The key difference from models 1 and 2 is that model 3 actively involves the creative
industries in the growth of the economy. This can occur in two principle ways: supply side
and demand side. The supply-side interpretation of this model emphasizes the export of new
ideas from CI to Y. The demand-side interpretation emphasizes how growth in Y causes a
proportionate increase in demand for CI services. In practice, it is extremely difficult to
separate these two forces without recourse to advanced microeconometric techniques,
which are not attempted here due to data limitations. Model 3 may therefore be true, but
with different policy implications depending upon whether causality runs predominantly
from CI to Y the supply-side growth driver model or from Y to CI the demand side
induction model.

238

J. Potts and S. Cunningham

Downloaded by [The Library, University of Witwatersrand] at 03:53 12 February 2015

Hypothesis 3:

dY
dU
> 0,
0
dCI
dCI

Yet in both cases, policy should properly treat the creative industries as a special
sector. This is not because it is economically significant in itself, but because it powers the
growth of other sectors. This may plausibly lead to intervention, but unlike model 1, the
ostensible purpose of this is to invest in economic growth, or to invest in the development
of capacity to meet growth in demand. If model 3 is true, then there is a clear economic case
for redirecting resources, not just for the benefit of the creative industries per se, but for the
benefit of all. The creative industries, in this view, are clear winners to be backed.
Evidence for this model would come from association of the creative industries with
growth. This would accrue not just in jobs and commodities (as in model 2), but in new
types of jobs and new sorts of commodities and services. Model 3 proposes the creative
industries as growth drivers not because of operational expenditure multipliers, but due to
their role in the adoption, retention and absorption of new ideas and technologies.
The creative industries would be thus assumed to create new industries and market
niches and to stabilize and develop extant industries. And specifically, without such continued investment, aggregate economic growth would suffer. This is the opposite of model 1,
in which economic growth suffers when there is such continued investment. Model 3 thus
argues that the creative industries are good for the economy because they introduce and
process the new ideas that drive economic growth. And that, correspondingly is their policy
significance, namely as an investment industry, just as coal and steel was in the late 1900s
and ICT was in the late twentieth century. The creative industries are a driver of growth,
and the more the better.
Model 4: The innovation model
These three models might seem exhaustive of analytic possibilities: yet a fourth model is
also possible. Rather than thinking of the creative industries as an economic subset driving
growth in the whole economy, as in model 3, the creative industries may not be well characterized as an industry per se, but rather as an element of the innovation system of the
whole economy.16
Model 4 hypothesizes the relationship between the creative industries and the rest of the
economy in a different way. Instead it is based on the contemporary innovation literature
sourced in the Schumpeterian tradition and applied mostly in the business and strategy literature (Metcalfe 1998). This model effectively rejects the initial statics-to-dynamics master
equation CI = A.Y and Yt/CIt. Instead, it re-conceptualizes the creative industries as a
higher-order system that operates on the economic system. Model 4 is similar to model 1,
in that it ventures an element of special pleading. Specifically, this is the same model as
proposed for the effect of science, education and technology in the national systems of innovation approach.17 The creative industries, in this view, originate and coordinate change in
the knowledge base of the economy. In consequence they have crucial, not marginal, policy
significance.
In model 4, the significance of the creative industries is not in terms of their relative
contribution to economic value (models 13), but due to their contribution to the coordination of new ideas or technologies, and thus to the process of change. In this view, the
creative industries are mis-specified as an industry per se, and better modeled as a complex
evolving system that derives its economic value from the facilitation of economic evolution and the process of innovation. The creative industries might in this sense be better

Downloaded by [The Library, University of Witwatersrand] at 03:53 12 February 2015

International Journal of Cultural Policy

239

understood as a kind of industrial entrepreneurship operating on the consumer side of the


economy (Potts et al. 2008). In this case, we are dealing with an evolutionary model of the
creative industries.
Change in the creative industries therefore produces structural and not just operational
change in the economy. For example, some of the most dramatic changes in contemporary
business models have been provoked by new uses of the internet in recent years (Anderson
2006). The culturization of the economy (Lash and Urry 1994) is now clear and is in
evidence in design-led innovation, the manifold industrial applications of games technologies and the impact of vernacular creativity and user-led innovations in mobile media use
(Cunningham 2006). New opportunities and possibilities will thus emerge of which the
welfare effect cannot be known in advance. This is the typical situation of economic evolution as the origination, adoption and retention of generic novelty (Dopfer and Potts 2008).
According to model 4, the creative industries do not drive economic growth directly, as
might a boom in the primary resource sector or the housing market for example, but rather
facilitate the conditions of change in the economic order. If model 4 is true, then, the
creative industries are part of the innovation system driving and coordinating the growth of
knowledge process that underpins economic evolution.18
Hypothesis 4:

dY
dU
undefined ,
open
dCI
dCI

Culture is indeed a public good, but for dynamic not static reasons. Unlike the value of
museums or classical arts, which seek cultural value through the maintenance of past
knowledge, creative industries value lies in the development and adoption of new knowledge. Evidence for model 4 therefore accrues from ongoing regeneration of existing industries and the emergence of new industries in consequence of creative industries activity.
Furthermore, this must be a systemic facilitator of ongoing structural change and adaptation across the whole economy. Model 4 thus requires observation of ongoing structural
change and re-consolidation across the entire economy, and with catalytic attribution of
this to creative industries operations. This is plainly difficult to test, yet this dynamic
distinction between growth processes (model 3) and evolutionary processes (model 4) is
important, for it carries the most radical policy implication: namely the possibility of a shift
of policy from an economic engagement with respect to welfare as opposed to an economic
engagement with respect to innovation.
In sum, these four models represent four possible modes of dynamic interaction between
the creative industries and the economy (CI = A.Y).

In model 1, Y drives CI through transfers of resources.


In model 2, the creative industries are just another industry.
In model 3, CI drives Y through high rates of growth.
In model 4, the creative industries evolve Y through transfers of knowledge.

Evidence
The substantial variation in performance measures within the creative industries cautions us
that what is statistically true of the aggregate is not necessarily true of sub-sectors. There is
substantial sub-sectoral variety in the creative industries both over time and in terms of their
business models.19 Clearly, the models discussed here accurately describe creative industries activity across the spectrum so that different models are more appropriate at different

240

J. Potts and S. Cunningham

Downloaded by [The Library, University of Witwatersrand] at 03:53 12 February 2015

times and places. We therefore emphasize that this is a first pass with a new methodology,
not a final conclusion.
The data necessary to discriminate between these models is at this stage very uneven.
Our preliminary sample of available data on differential growth in creative industries value
added predominately favours the prima facie conclusion that models 3 and 4 fit better than
models 1 or 2. The current evidence mostly points toward the creative industries growth
model.
Comparative growth evidence
There is a raft of recent aggregate data on the economic impact of the creative industries
from which we may infer relative growth rates. Recently, the data of creative industries has
gotten a lot better. The survey to follow may seem sketchy, but it is now orders of magnitude better than a decade ago, and now regularly updated. A current survey of data is by
definition out of date by the time published, and especially in this industry.
The basic finding for Australia, US, Britain, and the EU for 19992006 is that the
creative industries, under various definitions, are growing at a faster rate than the aggregate
economy. This is primary evidence of structural transformation as driven by the creative
industries, thus supporting model 3. Between 200005 the Australian creative industries
grew at twice the rate of the aggregate economy.20 The European Commission found that
the growth of the cultural and creative sector proceeded in Europe at 8%.21 In New Zealand,
it was found that creative industries value added has been recently growing at 8% per year.22
In the UK, where the most comprehensive data exists, the creative industries have been
recorded as growing at 5% as compared with real aggregate GDP growth of 3%. CI growth
ratios are everywhere greater than 1.0 (see Table 1).
The World Intellectual Property Organization (WIPO) estimates for the copyright based
(CB) industries show both value added and employment growth also at significantly higher
rates than aggregate GDP23 (see Table 2).
And although there are few comprehensive statistics dating back more than a decade,
Singapore data using the DCMS classification for compounded annual growth rate of
creative industries also indicates that this effect may not be a recent fluctuation, but part of
a sustained trend in post-industrial economies24 (see Table 3).
We may infer from this sample that the creative industries are now, and have been for
the past decade, growing at about twice the rate of the aggregate economy.25 In the absence
of evidence of increasing proportions of transfer payments to the creative industries, this
seems to reject models 1 and 2, and favour model 3.
Table 1.

Creative industries value-added and employment growth.

Country
Australia
20002005
New Zealand
19962001
Europe
19992003
UK 19972005

National
CI
CI value CI value
added GDP growth CI employment employment
added
growth
growth
(% GDP) growth growth ratio

CI
employment
growth ratio

6.0

10.4

4.0

2.6

3.8 19962001 1.9 19962001 2.0 19962001

3.1

3.7

2.2

1.6

2.6

5.4

2.9

1.9

na

na

7.3

5.0

3.0

1.7

2.0

1.0

International Journal of Cultural Policy


Table 2.

Copyright based industries value-added and employment growth.


National
CB
average
CB
CB value Average CB Average CB
GDP growth employment employment employment
value added
added
growth ratio
growth
growth
ratio
growth
growth
(% of GDP)

Downloaded by [The Library, University of Witwatersrand] at 03:53 12 February 2015

Country
Canada
19972002
USA
19992002
Singapore
19952000

Table 3.

241

5.4

6.5

3.3

1.9

5.3

1.4

3.8

11.9

2.4

1.0

2.4

2.0

1.4

1.4

5.7

8.9

7.6

1.2

5.2

3.5

1.5

Creative industries growth in Singapore 19862000.

Country
Singapore CI
Singapore GDP

198690

199095

199500

CI/GDP ratio

4.6
2.8

3.4
2.4

2.6
1.5

1.6

Firms
At the micro level, we may compare data on the growth in creative industries firms with
aggregate growth in all firms. Also, we may compare the profitability of creative industries
firms and aggregate profitability. According to economic theory, these statistics should be
related, as higher than average profitability would encourage a shift of resources into
creative industries, increasing the number of firms. In Australia, the creative industries
proportion of all Australian firms grew from 5.9% to 6.6% between 20002005. The total
number of creative industries firms grew at cumulative annualized growth rate (CAGR) of
11.3% between 20002005. For all Australian industries, the growth rate over the same
period was 8.3%.26 The sector has a higher rate of enterprise creation than the economy as
a whole, consistent with the observed growth rate of value added in the creative industries
sector.
Profit data is not widely reported by creative industries. Furthermore, it can be ambiguous, appearing low in both a declining industry due to low margins, as well as in a growing
industry due to reinvestment. European estimates of average profitability (return on capital
investment) of the cultural and creative industries for 19992003 was 9%, which is similar
to Australian estimates. This is good for the service sector, which in Europe ranges between
510%.27 This is an unsurprising result, signaling the creative industries as comparably
competitive, as model 2 hypothesizes. Similar profitability supports model 2, and enterprise
growth supports models 3 and 4. Yet acknowledging the considerable variation within the
creative industries, firm data only consistently rejects model 1.
Income
Income in the creative industries provides a good opportunity to discriminate between the
negative, neutral and positive models of creative industries. Recent Australian census data
indicates average income by six sectors for 2001.28 The all-industries mean was $36,276 and
was $47,658 for creative industries, which is 31% higher, although with significant variation
within this29 (see Table 4).

242

J. Potts and S. Cunningham

Table 4.

Australian mean income in CI sectors.

Downloaded by [The Library, University of Witwatersrand] at 03:53 12 February 2015

Australian mean income by sector 2001


software development and interactive content
film, TV, radio
advertising and marketing
writing, publishing and print media
architecture, design and visual arts
economy-wide average
music and performing arts

$64,288
$48,808
$48,278
$38,392
$37,658
$36,276
$32,553

There are several explanations for the higher creative industries income. First, the
creative industries have higher human capital than the aggregate economy.30 Second, it may
be that mean and median statistics do not accurately reflect the distribution of income,
which is instead heavily skewed to a winner-take-all situation. Indeed, there is substantial
evidence for a power law rather than Gaussian income distribution, in which the creative
industries are disproportionately represented among the super-rich (De Vany 2004, Potts
2007).31 A third possibility is that incomes are higher in the creative industries due to transfer payments from other sectors.32 This is dominated by heritage funding and public broadcasting, yet is broadly comparable with the EU and the US. Yet, to the extent these
constitute investment in social technologies the net transfer may even be from the creative
industries to the rest of the economy, which would be evidence against model 1 and for
model 4.
Implications
Considered together, this initial marshalling of evidence with which to test the four models
does not offer definitive conclusions, but does broadly tend to support models 3 and 4.
There are surely many further ways we could seek to develop and analyze data that might
test our four models of creative industries. However, there are good reasons we should
expect this manner of growth in the creative industries:
(1) rising affluence, which shifts aggregate expenditure toward the creative industries,
as their income elasticity is greater than unity;
(2) the related rise in human capital, which permits greater specialization;
(3) the growth in ICT, which is the technology base of the creative industries;
(4) globalization, as access to global markets both in demand and factor mobility.
The relative growth of the creative industries is not an anomaly, but what open-system
economic theory predicts as based on the effects of technological change (i.e. endogenous
growth) and a changed consumption set consistent with increased income.33 The evidence
we have surveyed broadly supports the model that creative industries growth is driving
economic evolution.
What does this imply for public policy? One perspective is to view the creative industries in the as the twenty-first century analogue of the creative destruction wrecking ball
that was nineteenth century engineering.34 The nineteenth and twentieth century transformations that wrought epoch making changes in the economy and culture occurred through
physical, chemical, civil and electrical engineering, along with economic engineering. The

Downloaded by [The Library, University of Witwatersrand] at 03:53 12 February 2015

International Journal of Cultural Policy

243

same argument now applies to creative industries engineering of open systems, rather than
closed systems. If this metaphorical translation holds, policy implications follow directly
from model 4 and with substantial and significant role for public support based on innovation policy. However, an equally consistent perspective may be read from model 3, in which
creative industries are growth mechanisms for the generic adjustment and adaptation of
the knowledge-base of the economy. In this case, policy has a distinctly less substantive
role: to minimize distortionary interference.
Models 3 and 4 thus harbour a substantially differential commitment to public intervention (model 4) or investment (model 3), which is why further work in empirically distinguishing these models, or in exploring whether they might be different aspects of a unified
model, is worth pursuing. We find, for example, that growth is the primary way that success
is spent in the creative industries (as opposed to higher rents or profits, which would be
expected in a static situation). Indeed, as DEMOS (2007) find, the creative industries grow
by staying small, such that growth occurs as spin-off entrepreneurial growth.35
As we noted in introducing the available evidence for the models, the creative industries
evidence substantial sub-sectorial diversity in business models and rates of growth in their
temporal and spatial dimensions. We should not therefore countenance any attempt to
impose a singular model on such diversity. However, it has been the purpose of this article
to indicate that new policy positions are required to address emergent realities around the
explosion of new mobile and internet media, design and user led innovation. These trends
clearly pose major challenges for the more established policy models. Having said that,
however, it is undeniable that in any given, real world, policy development process these
models may differentially apply.
This outline of policy implications suggests the need for new theory. Models 2 and 3
properly invite consistent treatment of the creative industries, but models 1 and 4 invite
substantial public intervention, although in different ways. Yet although our sample data
favours models 3 and 4 over models 1 and 2, it does not clearly distinguish between
models 3 and 4. The creative industries are therefore plausible drivers of economic
growth.36
The creative industries, in this view, have dynamic and not just static economic value
they contribute to the process of economic growth and development over and above their
contribution to culture and society. This distinction is important, as cultural policy, which
is traditionally based on model 1, may require some critical retooling to adapt to what
appears increasingly to be a model 4 world.
Notes
1. DCMS (1998), Cunningham (2001), Hesmondhalgh and Pratt (2005), Galloway and Dunlop

(2007).
2. For further discussions, see DCMS (2001), Howkins (2001), Florida (2002), Garnham (2005),

Hartley (2005), Cunningham (2006).


3. Creative industries is a wider definition than cultural industries that extends to include the osten-

sibly commercial fields of architecture, advertising, video games, software, and R&D. Floridas
(2002) occupation rather than industry-based classification goes further still to include all problem-solving work, thus further including broadly the white-collar as well as the no-collar workforce. See for example DCMS (1998, 2001), Higgs et al. (2007b).
4. See Howkins (2001), Garnham (2005), Cunningham (2006), Hesmondhalgh and Pratt (2005).
5. Strictly speaking, we are concerned not with absolute growth, but with distance-from-mean
dynamics (see Metcalfe 1998).
6. Any consistent operational measure will suffice, whether as an output such as gross domestic or
sectoral product (i.e. value-added), income or exports, or by measures of inputs such as employment,
capital (including human capital) and firm creation (entrepreneurship).

244

J. Potts and S. Cunningham

7. Which we may reasonably treat as parametric for a given endowment of resources, technologies,
8.
9.

Downloaded by [The Library, University of Witwatersrand] at 03:53 12 February 2015

10.

11.

12.
13.
14.
15.
16.
17.
18.
19.

20.

21.
22.

23.
24.
25.

26.

preferences and relative prices. Obviously, similar equations could be written for all other sectors,
subject to a unitary summation of scalars.
See Higgs and Cunningham (2007), Higgs et al. (2007a), NZIER (2002) and European Commission
(2006).
The notion that economic significance ipso facto implies proportional policy significance is a
widely-held fallacy that has been reproduced through generations of so-called economic impact
assessments. Economists have long cautioned against this interpretation (e.g. Seaman 1987, 2003).
Perhaps surprisingly, there are no empirical estimates for the income elasticity of the whole of
the cultural or creative industries, although based on estimates for elements of the creative
industries (e.g. demand for opera and theatre) they are broadly assumed to be greater than
unitary, and have been estimated to be so for public expenditure on culture (see Heilbrun and
Gray 2000, pp. 104106).
Note also that this would seem to suggest what economists call a production function of the
form Yt = f(CIt), with the models distinguished by whether is less than, equal to, or greater than
1. However, we do not seek to estimate a production function (and hence ) because of both
insufficient consistent time series data for CI and the difficulties of interpreting a cross-country
regression. Production functions for the cultural industries have been developed and estimated
(e.g. Gapinski 1980, 1984), Throsby 2006), but what we refer to here is the notion of an entire
industry as an input into a growth equation, which is less conventional.
See Galloway and Dunlop (2007).
See Higgs et al. (2007a) on incomes in the creative (not cultural) industries. See Cowan (1996)
for discussion of why productivity measures are misleading in the cultural/creative industries.
In which technological change is broadly defined to include not just origination of new ideas but
also their adoption and retention. See also Chai et al. (2007).
See Scott (2002, 2006).
A further variation on this theme of abandoning the concept of industry is to define the creative
industries as the space of economic activity in which markets and organizations are predominantly shaped by social networks (see Potts et al. 2008).
For example, Lundvall (1992), Nelson (1993, 2002), Freeman (1995), Edquist (1997).
See Loasby (1999) and Freeman (2002).
NESTA (2006) has recently identified four sub-sectors of the creative industries with different
business models: services (architecture, design, web-development); experiences (museums,
galleries, heritage, live music and performing arts); originals (visual art, crafts, antiques); content
(publishing, broadcasting games, film, recorded music).
There were 437,000 people employed in the Creative Industries segments in 2001 representing
5.4% of the Australian workforce. Almost $21 billion was generated during 2001 from the
salaries and wages of people employed in the creative segments representing 7% of the total
generated from all Australian employment. Between 19962001 the number of people in the core
Australian CI workforce has grown from approximately 150,000 to 180,000 a CAGR of 3.8%
(Higgs et al. 2007a).
The 5.4% average growth rate is for the EU 25. It was 8.1% for the total 30 European countries.
Prepared for Industry New Zealand. Estimates of the value added share of the copyright industries
for 1981 and 1986 were both at 3.0%, indicating that through the 1980s the copyright industries
were not growing faster than the economic aggregate average, but at the same rate. This is preliminary evidence that the creative industries in effect took-off during the 1990s. Over this same
period, employment in the CI grew by 5% per year. This resulted in the CI share of GDP growing
from 2.6% in 1996 to 3.2% in 2001 and employment from 3% to 3.6%.
Further supporting evidence of the relative growth in US employment in the creative industries
can be found in Florida (2002) and Levy and Murnane (2004). Note also that copyright based
industries are not identical to creative industries in that they also include distribution.
Economic contribution of Singapores creative industries Ministry of Trade and Industry,
Government of Singapore from the report Economic survey of Singapore first quarter 2003.
The creative industries are estimated to be proportionally smaller as a fraction of economic value
added in developing nations such as China, India, Russia and Brazil, yet their service sector
growth rates provide suggestive evidence that the relative growth ratios of the creative industries
may be similar to post-industrial societies. Much work remains to be done here.
wiki.cci.edu.au/display/NMP/Businesses+in+the+Creative+Industries (accessed 30 January
2008).

International Journal of Cultural Policy

245

27. The profit margin across all Australian CI firms was recently estimated at 10.1%. There was

28.
29.

Downloaded by [The Library, University of Witwatersrand] at 03:53 12 February 2015

30.

31.
32.

33.
34.
35.
36.

considerable variation between industries, from 4.8% in the performing arts to 24% in TV broadcasting. ABS data, reported in table 3.2 in CIRAC (2003).Yet the Australian creative industries
average falls about mid-way in the range of all Australian industries.
Conducted by CIRAC using Australian Tax Office data. This is a mean trident figure including
specialist creative, support staff and embedded creative.
Australian 2001 census data yields a similar conclusion, with the median weekly income in the
creative industries reported at $765, which is 15% higher than the all industries median of $663
(CIRAC, 2003).
European data reports that the cultural sector has higher human capital (46% with university
degrees) than in the aggregate economy (25% for all Europe). Similar results are reported for UK
and Australian creative industries. Sectoral aggregate qualifications and incomes exhibit positive
correlation (e.g. software development reports higher human capital measured by formal qualification than musicians).
For example, although composing about 6% of the Australian economy, creative industries
comprise 11% of Australias top 500 fortunes and 38% of those aged 40 and under. This pattern
repeats in the UK, USA and NZ.
Public funding of culture averages 0.51.0% of GDP for the EU, although with much variation,
as for example France and Austria spend close to 4%. US public funding is very low (less than
0.1%), but as Cowan (2006) explains, the figure for the total transfer is comparable to Europe
once trusts, not-for-profit organizations and corporate transfers are accounted for. In Australia,
government funding of cultural industries was $4 billion in 2001, which is about half of one
percent of GDP. Table 3.4 in CIRAC (2003) report.
See Potts and Mandeville (2007); Cowan (1998, 2002); Florida (2002); Howkins (2001).
John Hartley, personal communication (2006).
85% of CI firms employ fewer than five people, a percentage that has grown as the industry has
grown, implying entrepreneurship rather than scale as the dominant method of growth.
Nelson and Sampat (2001), Caves (2000), Dopfer and Potts (2008).

References
Anderson, C., 2006. The long tail. New York: Hyperion.
Baumol, W., and Bowen, W., 1966. Performing arts: The economic dilemma. New York: Twentieth
Century Fund.
Caves, R., 2000. Creative industries: Contracts between art and commerce. Cambridge, MA:
Harvard University Press.
Chai, A., Earl, P., and Potts, J., 2007. Fashion, growth and welfare: An evolutionary approach.
Advances in Austrian economics, 10, 187207.
CIRAC, Creative Industries Research and Application Centre, 2003. Brisbane creative industries
report. Brisbane: QUT.
Cowan, T., 1996. Why I do not believe in the cost-disease. Journal of cultural economics, 20, 207214.
Cowan, T., 1998. In praise of commercial culture. Cambridge, MA: Harvard University Press.
Cowan, T., 2002. Creative destruction. Princeton, NJ: Princeton University Press.
Cowan, T., 2006. Good and plenty. Princeton, NJ: Princeton University Press.
Cunningham, S., 2001. From cultural to creative industries: Theory, industry and policy implications. Culturelink, Special Issue, ed. Colin Mercer, 932.
Cunningham, S., 2006. What price a creative economy? Platform papers 9. Sydney: Currency
House.
DCMS, 1998. Creative industries mapping document. London: DCMS.
DCMS, 2001. Creative industries mapping document. London: DCMS.
DEMOS, 2007. So, what do you do? A new question for policy in a creative age. Green Paper.
De Vany, A., 2004. Hollywood economics. London: Routledge.
Dopfer, K., 2004. The economic agent as rule maker and rule user: Homo Sapiens Oeconomicus.
Journal of evolutionary economics, 14(2) 177195.
Dopfer, K., and Potts, J., 2008. The general theory of economic evolution. London: Routledge.
Edquist, C., 1997. Systems of innovation: Technologies, institutions and organizations. Washington
DC: Pinter.
Florida, R., 2002. The rise of the creative class. New York: Basic Books.

Downloaded by [The Library, University of Witwatersrand] at 03:53 12 February 2015

246

J. Potts and S. Cunningham

Freeman, C., 1995. The national system of innovation in historical perspective. Cambridge journal
of economics, 19, 524.
Freeman, C., 2002. Continental, national and sub-national systems of innovation. Research policy,
31, 191211.
Frey, B., 2003. Arts and economics: Analysis and cultural policy. Dsseldorf: Springer.
Galloway, S., and Dunlop, S., 2007. A critique of definition of the cultural and creative industries in
public policy. International journal of cultural policy, 13(1) 1731.
Gapinski, J.H., 1980. The production of culture. Review of economics and statistics, 62(4) 578586.
Gapinski, J.H., 1984. The economics of performing Shakespeare. American economic review, 74,
458466.
Garnham, N., 2005. From cultural to creative industries. International journal of cultural policy, 11,
1529.
Hartley, J., ed., 2005. Creative industries. Malden, MA: Blackwell.
Heilbrun, J., and Gray, C., 2000. The economics of art and culture. Cambridge: Cambridge University
Press.
Hesmondhalgh, D., and Pratt, A., 2005. Cultural industries and cultural policy. International journal
of cultural policy, 11, 113.
Higgs, P., and Cunningham, S., 2007. Australias creative economy: Mapping methodologies
[online]. Available from: http://eprints.qut.edu.au/archive/00006228/ [Accessed 30 January
2008].
Higgs, P, Cunningham, S., and Pagan, J., 2007a. Australias creative economy: Basic evidence on
size, growth, income and employment [online]. Technical Report, Faculty Research Office, CCI.
Available from: http://eprints.qut.edu.au/archive/00008241/ [Accessed 30 January 2008].
Higgs, P, Cunningham, S., and Pagan, J., 2007b. Australias creative economy: Definitions of the
segments and sectors. Technical Report, Faculty Research Office, CCI. Available from: http://
eprints.qut.edu.au/archive/00008242/ [Accessed 30 January 2008].
Howkins J., 2001. The creative economy. London: Allen Lane.
Lash, S., and Urry, J., 1994. The economics of signs and space. London: TCS Sage.
Levy, F., and Murnane, R., 2004. The new division of labour. Princeton, NJ: Princeton University
Press.
Loasby, B., 1999. Knowledge, institutions and evolution in economics. London: Routledge.
Lundvall, B.., 1992. National systems of innovation: Towards a theory of innovation and interactive learning. New York: Pinter Publishers.
Metcalfe, S., 1998. Evolutionary economics and creative destruction. London: Routledge.
NESTA, 2006. Creating growth: How the UK can create world class creative business. London:
NESTA.
Nelson, R., ed., 1993. National innovation systems: A comparative analysis. New York: Oxford
University Press.
Nelson, R., 2002. Technology, institutions and innovation systems. Research policy, 31, 265272
Nelson, R., and Sampat, B., 2001. Making sense of institutions as a factor shaping economic performance. Journal of economic behaviour and organization, 44(1) 3154.
Potts, J., 2007. How creative are the super-rich? Agenda, 13(4) 139150.
Potts, J., and Mandeville, T., 2007. Towards a theory of the evolution of services. Prometheus, 25(2)
147160.
Potts, J., Ormerod, P., Cunningham, S., and Hartley, J., 2008. Social network markets: A new definition of the creative industries. CCI working paper. Brisbane: QUT.
Scott, A., 2002. A new map of Hollywood: The production and distribution of American motion
pictures. Regional studies, 36(9) 957975.
Scott, A., 2006. Entrepreneurship, innovation and industrial development: Geography and the
creative field revisited. Small business economics, 26(1) 124.
Seaman, B., 1987. Arts impact studies: A fashionable excess. In: Economic impact of the arts: A
sourcebook. Washington DC: National Conference of State Legislatures, 4376.
Seaman, B., 2003. The economic impact of the arts. In: R. Towse, ed. The handbook of cultural
economics. Cheltenham: Edward Elgar.
Singapore, Government of, 2003. Economic survey of Singapore first quarter 2003. Singapore:
Ministry of Trade and Industry.
Throsby, D., 1994. The production and consumption of the arts. Journal of economic literature, 32,
129.

International Journal of Cultural Policy

247

Downloaded by [The Library, University of Witwatersrand] at 03:53 12 February 2015

Throsby, D., 2001. Economics and culture. Cambridge: Cambridge University Press.
Throsby, D., 2006. An artistic production function: Theory and an application to Australian visual
arts. Journal of cultural economics, 30, 114.
Throsby, D. and Withers, G., 1979. Economics of the performing arts. London: Edwards Arnold.
Towse, R., ed., 1997. Cultural economics. Cheltenham: Edward Elgar.
Towse, R., ed., 2003. A handbook of cultural economics. Cheltenham: Edward Elgar.